Celestia Forums

Is the model in Celestia accurate? It shows component A as a large G8III type star of 18 Rsun but the B component is only 1.2 Rsun, and the seperation is at 1.4 AU.

Burnhams has these stars much closer physically (70 million Miles) and much closer in size to each other, although the spectral classes are close to the book.

Other online sources also have the two stars being more alike.

http://www.solstation.com/stars2/capella4.htm

How does Celestia get the info for this system, is it a bit basic and for real accuracy each system would have to have a dedicated STC file?

I have also found other stars whose display in celestis do not seem to agree with the tycho catalogues specs.

Just wondering really, I enjoy celestia a lot but would like there to be more agreement between the display and other source info on the stars.

See this thread for some discussion of this system. The situation does not seem to be as clear-cut as some popular sources make out.

ajtribick wrote:See this thread for some discussion of this system. The situation does not seem to be as clear-cut as some popular sources make out.

Interesting, thanks for the link.

The image on the site I posted though seems to show the two seperated components as being fairly equal in mv, at least not so widely seperated in mv as displayed in the program ATM. One would imagine that a visula reference like that would at least carry a small ammount of weight in the case for a more balanced system as has been assumed by the likes of Burnham etc.

I was in error about the seperation though, on closer inspection that seems spot on, I made a mistake.

Kim,

Celestia uses the Hipparcos catalog as its primary source, translated into Celestia's binary file stars.dat. They're all defined as individual stars, not as multi-star systems. Corrections are applied by revised.stc. Some nearby stars which weren't measured by Hipparcos are defined in nearstars.stc, while additional stars known to be orbited by extrasolar planets are defined in extrasolar.stc. Celestia's binary stars are defined in spectbins.stc and visualbins.stc.

All of these stc files are plain text files in the /data/ directory. They include citations for the sources of their information.

Kim_Gowney wrote:Is the model in Celestia accurate? It shows component A as a large G8III type star of 18 Rsun but the B component is only 1.2 Rsun, and the seperation is at 1.4 AU.

Burnhams has these stars much closer physically (70 million Miles) and much closer in size to each other, although the spectral classes are close to the book.

Other online sources also have the two stars being more alike.

http://www.solstation.com/stars2/capella4.htm

How does Celestia get the info for this system, is it a bit basic and for real accuracy each system would have to have a dedicated STC file?


I have also found other stars whose display in celestis do not seem to agree with the tycho catalogues specs.

Just wondering really, I enjoy celestia a lot but would like there to be more agreement between the display and other source info on the stars.

In general our database (including notably galaxies and binaries) is prepared with outmost care and should represent the latest status of published knowledge (apart from possible bugs, of course)

Please consider that we need complete 3d orbit information about binary systems in Celestia. I have taken great care to extract the best available parameters from scientific publications with acclaimed reputation. All modiifications (including correction of typos) wrto the original publications are documented in human readable form in my respective perl scripts that are to be found in the Celestia->tools directory.

Some of the typical Capella problematics you may find discussed in this thread. Have a look at my replies! If you have a question, it is always worthwhile to first use this forum's powerful search engine for the 97221 articles!

http://www.celestiaproject.net/forum/viewtopic ... ht=capella

By looking to the top of my data files for binaries (visualbins.stc and spectbins.stc) you can easily trace and look up the respective scientific references.

Note that the radii of stars are NOT based on reliable sources up to now. In that respect I refer you to a new add-on by ajtribick based on latest interferometric measurements of 1944 star radii.

http://www.celestialmatters.org/?page_id=127
and the corresponding publication:

Richichi A., Percheron I., Khristoforva M. (2005) - ?€?CHARM2: an updated Catalog of High Angular Resolution Measurements?€

Thanks for the clarifications there Selden, I understand better now how the program is working.

As many of my observations include double stars I was particularly interested to them in the 3D environment of Celestia, in some cases this is possible, but I understand better now why many seem to be "Missing".

Nevertheless, Celestia is fantastic and gives a great insight into many aspects of the universe difficult to comprehend from the night sky and description alone, so cheers!!

cross posting above your last one ...

F.

Hi Fridger,
We must have posted at virtually the same instant
Currently reading and trying to grasp the significance of the various posts and links etc in that thread. I very much appreciate your advanced understanding of the subject as I am heving to read and re-read this stuff several times over to try and get a handle on what is the basic influence on your desicion to render the B component as you have.

I am very interested, but not particularly academicaly gifted so a lot of this stuff has really gone voom, straight over the top!

I have full confidence though that you do understand very well, and also that the accuracy of celestia's displays are the best you can get given the available info etc, but if I were to ask one more question on this (as I currently am understanding it) why, if the nature and Abs of Ab is such an unknown did you elect to display it so differently from what seems to be the mainstream views (albeit they may be totally innacurate), would it not be just as arbitrary to assign it a value/radius nearer Aa? Innaccurate perhaps and yet somewhat less puzzling to the layman/men/women such as I?

Please do not become irritated if I have displayed a lack of understanding here, I intend no critisism, I thought I saw an error and am merely engaged in the process of discovery.
I appreciate your taking time to reply.

Kim_Gowney wrote:
I am very interested, but not particularly academicaly gifted so a lot of this stuff has really gone voom, straight over the top!

Please do not become irritated if I have displayed a lack of understanding here, I intend no critisism, I thought I saw an error and am merely engaged in the process of discovery.
I appreciate your taking time to reply.

Don't worry yourself too much there Kim... Doctor Schrempp is used
to Brain-Dead Geezers interrupting his chain of thought. Or, in his
case, CHAINS (plural).

If you think he's wrong, give him Hell. Chances are VERY slim that he
is wrong though. This from the voice of uncomfortable experience.
Hee, hee. After all, we can't have him thinking that he's perfect
now, can we?

Kim_Gowney wrote:I have full confidence though that you do understand very well, and also that the accuracy of celestia's displays are the best you can get given the available info etc, but if I were to ask one more question on this (as I currently am understanding it) why, if the nature and Abs of Ab is such an unknown did you elect to display it so differently from what seems to be the mainstream views (albeit they may be totally innacurate), would it not be just as arbitrary to assign it a value/radius nearer Aa? Innaccurate perhaps and yet somewhat less puzzling to the layman/men/women such as I?

The binary files are derived from quite large datasets, so going off chasing every system to "fill in the blanks" is a time-consuming pursuit, and would in any case have the disadvantage that it starts mixing up data sources. The generation is done via scripting, which means it doesn't know what the "mainstream views" are.

If you want to use the solution given in Hummel et al. 1994, including distance and radii for the two components create the following stc file:

EDIT: code has been updated, see later in thread

NOTE 1: I haven't tested this code, because I'm using the OS without Celestia installed. (Waiting for 1.5.0 before I install on Linux)

NOTE 2: Even if it does work, my CHARM2 add-on will mangle the A star unless the code gets loaded AFTER "charm2.stc", so name it something like "hummel.stc" or anything which comes later alphabetically than charm2.stc.

Thanks Andrew.

NB:
I think, Andrew's radii should anyway become part of the Celestia distribution!

F.

BobHegwood wrote:
Don't worry yourself too much there Kim... Doctor Schrempp is used
to Brain-Dead Geezers interrupting his chain of thought. Or, in his
case, CHAINS (plural).

Hey hey, Bob,

don't forget...you are a special "Brain Dead geezer"

Cheers,
Fridger

NB: Kim, let me tell you in private, he's Einstein's nephew!

Wow! I made the stc and the result is fantastic! many thanks Andrew, I am very grateful.

I do note that both sets of stars are now displayed, is there a simple way to cancel the original stars?

Also to Fridger too for his input and explanaitions. I understand a lot better now why things are as they are, but it's wonderful that details can be brought in in this manner.

I did notice in another thread on binaries that Andrew contibuted similar data that the stars were showing orbital rings, can I get that display for this system? (I don't see an option for it in the display controls)


Again, many thanks to all for your help and input on this question, it has been both intersting and instructive.

Blimey... you guys... hehe... great stuff!

t00fri wrote:Thanks Andrew.

NB:
I think, Andrew's radii should anyway become part of the Celestia distribution!

F.

I agree completely. I'm preparing a (short) list of features for 1.5.1; I will add this.

--Chris

Kim_Gowney wrote:I do note that both sets of stars are now displayed, is there a simple way to cancel the original stars?

Hmmm evidently overloading only works if the set of star names is EXACTLY the same... my mistake. Try removing the :Capella A and :Capella B strings, so it now reads...

EDIT: Code had incorrect RA

Does that work?

t00fri wrote:NB:
I think, Andrew's radii should anyway become part of the Celestia distribution!

That'd be nice. (Hmmm... this is starting to look like a trend with my add-ons...)

Ideally to minimise interactions (e.g. if one of the other data files gets updated to change distance, binary properties, etc.) it would be better to have Modify directives and a way of specifying angular diameters in the .stc file: at present the add-on has to replicate lots of data that's already contained in the existing datafiles.

The other major mistake I made was that I forgot to convert the RA from hours to degrees, so that .stc file puts the star in the wrong place. The orbital elements on the other hand did take this into account.

In any case, the corrected version reads:

Thanks Andrew, (I had figured that one out )